Monday, February 22, 2016

January 2016 CPI

Well, I'd say this is a good development.  It looks less likely now that we will be tricked by high shelter inflation into a deflationary shock for the rest of the economy.  Non-shelter core inflation seems to be trending up now.

Here are my graphs for core inflation, with and without shelter, updated.  I don't think we are totally out of the woods.  In some ways, I think we are basically in a similar position as late 2007, but without the housing boom.  Then the Fed appeared to be loosening, and inflation recovered while the economy appeared to be stabilizing.  But, by the end of 2008, we discovered that things weren't nearly as stable as we had thought.

I think something like that is still a possibility.  It is still the case, as it has been for the better part of 20 years, that inflation is being pushed up by supply constraints in the housing market.  This will still cause the Fed to tighten monetary policy to a level below their stated targets.  Since 1997, non-shelter inflation has really fluctuated around a stable trend of about 1.5%, generally moving between 1% and 2%.  But, the Fed stance has shifted to a more hawkish posture, which seems to treat 2% as an inflation ceiling.  If shelter inflation remains above 3%, Core minus Shelter inflation may range even lower as we move forward.

I think the market tends to share this concern.  Friday, the Eurodollar yield curve flattened somewhat on the news.  The short end of the curve moved up about 3 basis points, but the long end of the curve moved down up to 5 basis points.  My model had shown an expected second rate hike as late as the end of 2017, last week.  This has moved back to about April 2017 now.  But, the slope of the curve after the hike continues to flatten.  It is so low that now, after the expected rate hike of April 2017, rates are expected to rise by only 1/4% per year!  This is essentially a flat yield curve.  I hope there is enough strength in the economy for recovery to continue.  Recent dovish comments from some Fed members are heartening.

Here are some long term graphs of shelter inflation, by region.  The two Americas come across pretty clearly here.  The first graph is the ratio of Shelter price levels to the price level of Core CPI less Shelter.  Rent prices in the Midwest and the South have risen fairly closely to Core CPI prices.

The Northeast has had unusual rent inflation since the late 1980s.  Rent inflation was pretty normal in the West, along with the South and the Midwest, until the mid-1990s.  This is when the urban housing shortage kicked into gear.  Since the mid-1990s, rents in the West have grown at a faster rate and have been more volatile than the other regions.

The last graph shows Trailing 12 month Shelter inflation for each region.  Here we can see that rent inflation shot up in all regions after housing starts collapsed in 2006, though the Midwest didn't rise quite as sharply.  Regional rent inflation tends to move somewhat independently.  But, at the beginning of 2006, there was a singular outside force that caused rent inflation to shoot up across regions at a similar scale.  A sharp negative shock in credit markets caused a sharp negative shock in housing supply.  Then, by mid-2007, the credit crisis had begun to drive home prices down, leading to defaults and foreclosures.  By then, households were sharply reducing their housing demand because they were literally losing their houses, and moving into whatever unit they could manage.  This sharp decline in rent inflation, and the initial recovery from it, were also both quite uniform.

Now, we are going back to a context where regional differences are more important, and rents in the West are shooting up more than in the other regions, although our continued War on Credit means that there is a lack of supply in all regions, so rent inflation is high in all regions.


  1. Since reading your blog I have come to realize that the CPI core is much less important than the shelter component.

    Shelter or housing is something close to 40% of the CPI. Energy and food must be around 22%.

    You really struck a chord with your manufactured housing post. After all, the price of a new car in the United States has not changed in 20 years. I am talking nominally! Imagine if the manufactured housing industry Washington allowed to place shelter wherever a buyer wanted it.

    1. Yes. It's a facet to the story that I have not addressed as much as I need to. I hope to have another post soon. It's highly concentrated in the south. In fact, I think this may be another sort of 2 Americas issue. High rate subprime type loans in rural areas largely fund manufactured homes for low income households. In the problem cities, they fund homes for middle income households where rents are too high.

      You won't be surprised to hear that the New England, Pacific coast, and mid-Atlantic states, combined, only received about 13% of US manufactured home shipments in 2014, even though they hold about 35% of the population.

  2. Mass manufacture of homes would allow much easier enforcement of durability/environmental/infrastructural compatibility standards. High-efficiency gas furnaces and thick insulation are easier to bundle than to mandate a la carte. Stress suffered by the non-expert masses in dealings with tradesmen and local bureaucrats is probably a serious drag on public health, in and of itself.

    Construction productivity is stagnant and we can't get enough workers. We would respond better to positive demand shocks, like visa reform. Notice that the European country with the "right to build" is the most zealous grantor of succor to refugees

    1. Interesting points. This definitely seems like a context where federalism fails. It's too bad the commerce clause has been neutered, because you're right.

  3. I'd definitely prefer single-planner, so to speak, like they have in Japan

    similarly, I'm sympathetic to Bernie Sanders style healthcare prescriptions because I think healthcare will always b political, and that nothing less than a Volcker-style central bureaucrAtlas can defeat the doctors and nurses - that or a more liberal visa regime, which a non-suicidal country would seem to want for all kinds of reasons

    1. Yeah. In education, health care, and housing, we have policies that arbitrarily constrict supply (although much of education supply constriction is just a natural outgrowth of a status-based service) then we try to solve that problem with massive subsidization of wasteful demand. This is a broad and obvious problem, that in a lot of ways would be quite easy to solve if political institutions were in the least bit functional. So, I haven't put much thought into the different health care systems. You don't play William Tell with a shotgun.